DESCRIPTION: This competitive renewal application addresses structural and molecular recognition issues associated with peptide-RNA interactions critical to the survival and propagation of oncogenic and immunodeficiency viruses, and the structural consequences of DNA damage that can trigger oncogenic transformations. The NMR based projects are directed towards an understanding of the principles, patterns and diversity associated with peptide-RNA recognition range from the molecular characterization of adaptive binding of the arginine-rich Rex peptide of HTLV-1 bound to the major groove of its RxRE RNA and RNA aptamer targets, and to the P22 N-peptide bound to the major groove of its boxB RNA target associated with the onset of transcriptional antitermination. In addition, it is planned to elucidate within the framework of the HIV-1 Rev peptide-RRE RNA complex how two different peptides can target the same RNA binding site with comparable affinity, as well as how the same peptide can target two distinct RNA binding sites. These structural and recognition studies are being extended to peptides that target the minor groove of the tRNAA1a acceptor microhelix with exquisite discrimination associated with the alignment directionality of an exocyclic amino group of a specificity-determining G*U mismatch pair. The ongoing NMR based tumorigenic DNA damage structural research will be extended to hot spot sequences, where the aromatic amine-C8-guanine adducts are positioned within runs of guanines, at specific guanines within the NarI sequence context, and at junctional positions within model template-primer junctions. These studies will probe the role of base sequence and adduct size on the distribution of base pair substitution and frameshift mutations. Recent cancer genetics research on Ashkenazi Jewish families have identified a high incidence of founding mutations at hot spot sites in the colorectal tumor APC (3920 T to A transversion) gene and the breast cancer BRCA1 (185de1AG) and BRCA2 (617delT) genes. A series of experiments are proposed toward a molecular hot spot. Mutations initially occurred as a result of adduct induced damage of these DNA sites. The DNA-binding domain of the 14.7 kD human nucleotide excision repair XPAC protein will be overexpressed, purified, and isotopically labeled for complexation studies with adduct containing duplexes. Such studies should provide insights into the recognition and discrimination of damaged substrates by the repair machinery.